• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.4
• /
• pp.580-585
• /
• 2009

This paper introduces GP(Genetic Programming) based corner detectors for an image processing. Various empirical algorithms have been studied to improve computational speed and accuracy including typical approaches, such as Harris and SUSAN. The these techniques are highly efficient, because properties of corner points are inspected and reflected into the algorithms. However these approaches are limited in discovering an innovative algorithm. In this study, we try to discover a more efficient technique by creating corner detector automatically using evolution of GP. The proposed method is compared to the existing corner detectors for test images.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.44 no.6
• /
• pp.468-476
• /
• 2016

In this study, the conic winglet which is made by rotating wing tip airfoil by each 3 axis is applied to the dual type combined fan to reduce the wing tip leakage loss. Computational Fluid Dynamics is used to calculate the loss and optimum technique is used to get minimum loss. Optimization results shows that total pressure loss coefficient was reduced by 3.4 %, and optimization model was a bended shape at the end of wing forward to pressure side.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.6 no.4
• /
• pp.76-85
• /
• 1998

Recently, Genetic Algorithm(GA) is widely adopted into a search procedure for structural optimization, which is a stochastic direct search strategy that mimics the process of genetic evolution. This methods consist of three genetics operations maned selection, crossover and mutation. Contrast to traditional optimal design techniques which use design sensitivity analysis results, GA, being zero-order method, is very simple. So, they can be easily applicable to wide area of design optimization problems. Also, owing to multi-point search procedure, they have higher probability of converge to global optimum compared to traditional techniques which take one-point search method. In this study, a method of finding the optimum values of suspension parameters is proposed by using the GA. And vehicle is modelled as planar vehicle having 5 degree-of-freedom. The generalized coordinates are vertical motion of passenger seat, sprung mass and front and rear unsprung mass and rotate(pitch) motion of sprung mass. For rapid converge and precluding local optimum, share function which distribute chromosomes over design bound is introduced. Elitist survival model, remainder stochastic sampling without replacement method, multi-point crossover method are adopted. In the sight of the improvement of ride comfort, good result can be obtained in 5-D.O.F. vehicle model by using GA.

• The Bulletin of The Korean Astronomical Society
• /
• v.36 no.2
• /
• pp.64-64
• /
• 2011

We present a new quasi-stellar object (QSO) selection algorithm using a Support Vector Machine, a supervised classification method, on a set of extracted time series features including period, amplitude, color, and autocorrelation value. We train a model that separates QSOs from variable stars, non-variable stars, and microlensing events using 58 known QSOs, 1629 variable stars, and 4288 non-variables in the MAssive Compact Halo Object (MACHO) database as a training set. To estimate the efficiency and the accuracy of the model, we perform a cross-validation test using the training set. The test shows that the model correctly identifies ~80% of known QSOs with a 25% false-positive rate. The majority of the false positives are Be stars. We applied the trained model to the MACHO Large Magellanic Cloud (LMC) data set, which consists of 40 million lightcurves, and found 1620 QSO candidates. During the selection, none of the 33,242 known MACHO variables were misclassified as QSO candidates. In order to estimate the true false-positive rate, we crossmatched the candidates with astronomical catalogs including the Spitzer Surveying the Agents of a Galaxy's Evolution (SAGE) LMC catalog and a few X-ray catalogs. The results further suggest that the majority of the candidates, more than 70%, are QSOs.

• The Bulletin of The Korean Astronomical Society
• /
• v.36 no.2
• /
• pp.53.2-53.2
• /
• 2011

We combine the physics of the ellipsoidal collapse model with the excursion set theory to study the shapes of dark matter halos. In particular, we develop an analytic approximation to the nonlinear evolution that is more accurate than the Zeldovich approximation; we introduce a planar representation of halo axis ratios, which allows a concise and intuitive description of the dynamics of collapsing regions and allows one to relate the final shape of a halo to its initial shape; we provide simple physical explanations for some empirical fitting formulae obtained from numerical studies. Comparison with simulations is challenging, as there is no agreement about how to define a non-spherical gravitationally bound object. Nevertheless, we find that our model matches the conditional minor-to-intermediate axis ratio distribution rather well, although it disagrees with the numerical results in reproducing the minor-to-major axis ratio distribution. In particular, the mass dependence of the minor-to-major axis distribution appears to be the opposite to what is found in many previous numerical studies, where low-mass halos are preferentially more spherical than high-mass halos. In our model, the high-mass halos are predicted to be more spherical, consistent with results based on a more recent and elaborate halo finding algorithm, and with observations of the mass dependence of the shapes of early-type galaxies. We suggest that some of the disagreement with some previous numerical studies may be alleviated if we consider only isolated halos.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.3 no.4
• /
• pp.263-273
• /
• 2011

The diffuse interface model of Saurel et al. (2008) is used for the computation of compressible cavitating flows around underwater missiles. Such systems use gas injection and natural cavitation to reduce drag effects. Consequently material interfaces appear separating liquid and gas. These interfaces may have a really complex dynamics such that only a few formulations are able to predict their evolution. Contrarily to front tracking or interface reconstruction method the interfaces are computed as diffused numerical zones, that are captured in a routinely manner, as is done usually with gas dynamics solvers for shocks and contact discontinuity. With the present approach, a single set of partial differential equations is solved everywhere, with a single numerical scheme. This leads to very efficient solvers. The algorithm derived in Saurel et al. (2009) is used to compute cavitation pockets around solid bodies. It is first validated against experiments done in cavitation tunnel at CNU. Then it is used to compute flows around high speed underwater systems (Shkval-like missile). Performance data are then computed showing method ability to predict forces acting on the system.

• The Bulletin of The Korean Astronomical Society
• /
• v.38 no.1
• /
• pp.42.1-42.1
• /
• 2013

In the LCDM paradigm, hierarchical merging is thought to play a key role in the formation and evolution of massive galaxies. Theoretical and observational studies suggest that massive galaxies started forming at high redshifts and were assembled via numerous mergers. Galaxy clusters are the sites where the most massive galaxies are found and the most dramatic merger histories are embedded. The previous work of Sheen et al. (2012) identified via visual inspection many massive galaxies with merger features in clusters, which surprised the community. In this study we aim to quantify peculiarity of galaxies to pin down the merger frequency in cluster environments more objectively. We have performed optical deep imaging of 4 Abell clusters by using IMACS f/2 on a Magellan Badde 6.5-m telescope. For the galaxies in our data, we applied GALFIT algorithm, which fits analytic models to galaxy data, and we analyzed their residuals. We present the preliminary results of our sample galaxies.

• The Bulletin of The Korean Astronomical Society
• /
• v.39 no.1
• /
• pp.33.2-33.2
• /
• 2014

Halo merger trees are essential backbones of semi-analytic models for galaxy formation and evolution. Recent studies have pointed out that extracting merger trees from numerical simulations of structure formation is non-trivial; different algorithm can give differing merger histories. Thus they should be carefully understood before being used as input for models of galaxy formation. As one of the projects proposed in the SUSSING MERGER TREES Workshop, we investigate the impact of different halo merger trees on a semi-analytic model. We find that the z = 0 global galaxy properties in our model show differences between trees when using a common parameter set, but that these differences are not very significant. However, the star formation history of the Universe and the properties of satellite galaxies can show marked differences between trees with different methods for constructing a tree. Calibrating the SAM for each tree individually to the empirical data can reduce the discrepancies between the z = 0 global galaxy properties, however this is at cost of increasing the differences in evolutionary histories of galaxies. Furthermore, the underlying physics implied can vary, resulting in key quantities such as the supernova feedback efficiency differing by factors of 2. Such a change alters the regimes where star formation is primarily suppressed by supernovae. Therefore, halo merger trees extracted from a common halo catalogue using different, but reliable, algorithms can result in a difference in the semi-analytic model, however, given the enormous uncertainties in galaxy formation physics, these are not necessarily significant.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.3
• /
• pp.792-801
• /
• 1997

In this paper we propose an algorithm which fits a surface to noise-corrupted data points using an aceive sur-vace with global deformations.It is 3-dimensional surface extension of our precious works on 2-dimensional curve modell[11,12].We use fimite differences,and represent shapes of surface as global transfromation suring evolution based on the ballon [odel[2,3]which allows partial inflation or deflation by using niegborhood'sextermal foreces on each node points.At first,we make local deformations based on the balloon [odel,and then the blobal transformation from the surface before deformations to the defomed surface is calcualted by the deterministic lest squre method,finally we apply the global transformation to the surface before deformations.In experiments we fit a surface(elliposid,B-spline)to noise-corrupted 3D data points using the active surface with affine deformations.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.24 no.5
• /
• pp.733-743
• /
• 2000

We solve the integral representation of the Navier-Stokes equations in a lagrangian view by tracking the particles, which have vortex strengths. We simulate the unsteady viscous flow around an impulsively started cylinder using the vortex particle method. Particles are advanced via the Biot-Savart law for a lagrangian evolution of particles. The particle strength is modified based on the scheme of particle strength exchange. The solid boundary satisfies the no-slip boundary condition by the vorticity generation algorithm. We newly modify the diffusion scheme and the boundary condition for simulating an unsteady flow efficiently. To save the computation time, we propose the mixed scheme of particle strength exchange and core expansion. We also use a lot of panels to ignore the curvature of the cylinder, and not to solve the evaluation of the surface density. Results are compared to those from other theoretical and experimental works.